Title :
Skill acquiring in robotics by successive approximation of nominal control sequences
Author :
Djordjevic, Goran S. ; Potkonjak, Veljko
Author_Institution :
Fac. of Electron. Eng., Nish Univ., Yugoslavia
Abstract :
Skill acquiring process for intelligent planning and trajectory following tasks based on learning is a topic considered in this paper. Several problems like definition and recognition of a skill, skill parametrization, off-line training, storage and generalization of a skill are particularly discussed. Task-oriented anthropomorphic learning concept, as a final result, gains a parametrized model of a robot for the given task. Successive approximation of nominal control sequences of a robot is used for experience generation. This approach possesses characteristics of globality. Applied in feedforward control concept it provides anticipation of future actions in trajectory planning tasks
Keywords :
generalisation (artificial intelligence); intelligent control; knowledge acquisition; learning (artificial intelligence); path planning; robots; tracking; anthropomorphic learning; intelligent planning; nominal control sequences; robotics; skill acquiring; skill generalization; skill parametrization; successive approximation; trajectory following; trajectory planning; Anthropomorphism; Humans; Intelligent robots; Intelligent sensors; Kinematics; Motion planning; Robot control; Robot sensing systems; Shape control; Shape measurement;
Conference_Titel :
Systems, Man and Cybernetics, 1995. Intelligent Systems for the 21st Century., IEEE International Conference on
Conference_Location :
Vancouver, BC
Print_ISBN :
0-7803-2559-1
DOI :
10.1109/ICSMC.1995.538363
